Generation of Action Recognition Training Data Through Rotoscoping and Augmentation of Synthetic Animations.
Nicola CovreFabrizio NunnariAlberto FornaserMariolino De CeccoPublished in: AVR (2) (2019)
Keyphrases
- action recognition
- training data
- bag of words
- human actions
- computer vision
- human detection
- motion capture data
- activity recognition
- spatial temporal
- decision trees
- action classification
- recognition of human actions
- static images
- learning algorithm
- view invariant
- action primitives
- training set
- action recognition in videos
- recognizing actions
- body parts
- computer graphics
- bag of features
- human activities
- labeled data
- supervised learning
- video dataset
- depth sensors
- text classification
- three dimensional
- independent subspace analysis